Hugging Face's logo

Datasets:
optimum-benchmark
/
rocm

Dataset card Viewer Files Community
IlyasMoutawwakil HF staff commited on
Commit
b106be0
·
verified ·
1 Parent(s): ea078aa

Upload cuda_training_transformers_multiple-choice_FacebookAI/roberta-base/benchmark_report.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. cuda_training_transformers_multiple-choice_FacebookAI/roberta-base/benchmark_report.json +43 -43
cuda_training_transformers_multiple-choice_FacebookAI/roberta-base/benchmark_report.json CHANGED
@@ -2,34 +2,34 @@
2
  "overall": {
3
  "memory": {
4
  "unit": "MB",
5
- "max_ram": 1332.3264,
6
  "max_global_vram": 68702.69952,
7
- "max_process_vram": 297091.710976,
8
  "max_reserved": 2707.423232,
9
  "max_allocated": 2497.88416
10
  },
11
  "latency": {
12
  "unit": "s",
13
  "values": [
14
- 0.6690802612304687,
15
- 0.0466182861328125,
16
- 0.045612041473388674,
17
- 0.04551459884643555,
18
- 0.045711402893066404
19
  ],
20
  "count": 5,
21
- "total": 0.8525365905761718,
22
- "mean": 0.17050731811523437,
23
- "p50": 0.045711402893066404,
24
- "p90": 0.4200954711914063,
25
- "p95": 0.5445878662109374,
26
- "p99": 0.6441817822265624,
27
- "stdev": 0.2492867835670069,
28
- "stdev_": 146.20298197320236
29
  },
30
  "throughput": {
31
  "unit": "samples/s",
32
- "value": 58.64850911115543
33
  },
34
  "energy": null,
35
  "efficiency": null
@@ -37,31 +37,31 @@
37
  "warmup": {
38
  "memory": {
39
  "unit": "MB",
40
- "max_ram": 1332.3264,
41
  "max_global_vram": 68702.69952,
42
- "max_process_vram": 297091.710976,
43
  "max_reserved": 2707.423232,
44
  "max_allocated": 2497.88416
45
  },
46
  "latency": {
47
  "unit": "s",
48
  "values": [
49
- 0.6690802612304687,
50
- 0.0466182861328125
51
  ],
52
  "count": 2,
53
- "total": 0.7156985473632812,
54
- "mean": 0.3578492736816406,
55
- "p50": 0.3578492736816406,
56
- "p90": 0.6068340637207031,
57
- "p95": 0.6379571624755859,
58
- "p99": 0.6628556414794922,
59
- "stdev": 0.3112309875488281,
60
- "stdev_": 86.9726475470546
61
  },
62
  "throughput": {
63
  "unit": "samples/s",
64
- "value": 11.177890509171709
65
  },
66
  "energy": null,
67
  "efficiency": null
@@ -69,32 +69,32 @@
69
  "train": {
70
  "memory": {
71
  "unit": "MB",
72
- "max_ram": 1332.3264,
73
  "max_global_vram": 68702.69952,
74
- "max_process_vram": 297091.710976,
75
  "max_reserved": 2707.423232,
76
  "max_allocated": 2497.88416
77
  },
78
  "latency": {
79
  "unit": "s",
80
  "values": [
81
- 0.045612041473388674,
82
- 0.04551459884643555,
83
- 0.045711402893066404
84
  ],
85
  "count": 3,
86
- "total": 0.13683804321289064,
87
- "mean": 0.04561268107096355,
88
- "p50": 0.045612041473388674,
89
- "p90": 0.04569153060913086,
90
- "p95": 0.04570146675109863,
91
- "p99": 0.04570941566467285,
92
- "stdev": 8.034618848608875e-05,
93
- "stdev_": 0.17614879590411997
94
  },
95
  "throughput": {
96
  "unit": "samples/s",
97
- "value": 131.54236627014507
98
  },
99
  "energy": null,
100
  "efficiency": null
 
2
  "overall": {
3
  "memory": {
4
  "unit": "MB",
5
+ "max_ram": 1338.667008,
6
  "max_global_vram": 68702.69952,
7
+ "max_process_vram": 223054.41792,
8
  "max_reserved": 2707.423232,
9
  "max_allocated": 2497.88416
10
  },
11
  "latency": {
12
  "unit": "s",
13
  "values": [
14
+ 0.591154052734375,
15
+ 0.044821876525878906,
16
+ 0.04357053375244141,
17
+ 0.04355213165283203,
18
+ 0.043511333465576174
19
  ],
20
  "count": 5,
21
+ "total": 0.7666099281311034,
22
+ "mean": 0.15332198562622068,
23
+ "p50": 0.04357053375244141,
24
+ "p90": 0.3726211822509766,
25
+ "p95": 0.4818876174926757,
26
+ "p99": 0.5693007656860352,
27
+ "stdev": 0.21891659325901863,
28
+ "stdev_": 142.78225811183347
29
  },
30
  "throughput": {
31
  "unit": "samples/s",
32
+ "value": 65.22221819106045
33
  },
34
  "energy": null,
35
  "efficiency": null
 
37
  "warmup": {
38
  "memory": {
39
  "unit": "MB",
40
+ "max_ram": 1338.667008,
41
  "max_global_vram": 68702.69952,
42
+ "max_process_vram": 223054.41792,
43
  "max_reserved": 2707.423232,
44
  "max_allocated": 2497.88416
45
  },
46
  "latency": {
47
  "unit": "s",
48
  "values": [
49
+ 0.591154052734375,
50
+ 0.044821876525878906
51
  ],
52
  "count": 2,
53
+ "total": 0.6359759292602539,
54
+ "mean": 0.31798796463012696,
55
+ "p50": 0.31798796463012696,
56
+ "p90": 0.5365208351135254,
57
+ "p95": 0.5638374439239502,
58
+ "p99": 0.5856907309722901,
59
+ "stdev": 0.27316608810424803,
60
+ "stdev_": 85.90453680282704
61
  },
62
  "throughput": {
63
  "unit": "samples/s",
64
+ "value": 12.579092434056953
65
  },
66
  "energy": null,
67
  "efficiency": null
 
69
  "train": {
70
  "memory": {
71
  "unit": "MB",
72
+ "max_ram": 1338.667008,
73
  "max_global_vram": 68702.69952,
74
+ "max_process_vram": 223054.41792,
75
  "max_reserved": 2707.423232,
76
  "max_allocated": 2497.88416
77
  },
78
  "latency": {
79
  "unit": "s",
80
  "values": [
81
+ 0.04357053375244141,
82
+ 0.04355213165283203,
83
+ 0.043511333465576174
84
  ],
85
  "count": 3,
86
+ "total": 0.13063399887084962,
87
+ "mean": 0.0435446662902832,
88
+ "p50": 0.04355213165283203,
89
+ "p90": 0.043566853332519535,
90
+ "p95": 0.04356869354248047,
91
+ "p99": 0.04357016571044922,
92
+ "stdev": 2.4738192061676743e-05,
93
+ "stdev_": 0.05681107278848744
94
  },
95
  "throughput": {
96
  "unit": "samples/s",
97
+ "value": 137.7895506191736
98
  },
99
  "energy": null,
100
  "efficiency": null